metadata
{}
MMMLU Dataset (merged across languages)
This dataset is a merged version of the MMMLU dataset across candidate languages: IT_IT, ES_LA, ZH_CN, HI_IN and the Russian MMMLU dataset from NLPCoreTeam/mmlu_ru.
Original Datasets
The original MMMLU dataset can be found here. Additionally, we include the Russian MMMLU dataset from NLPCoreTeam/mmlu_ru.